zlacker

ANNs are just mathematical transformations, powered by linear algebra + non-linear functions. They simulate certain cognitive processes — but they are fundamentally math, not magic.

>>cuttot+(OP)
I wouldn't say they "simulate cognitive processes". They do statistics. Advanced multivariadic statistics.

An LLM thinks in the same way excel thinks when you ask it to fit a curve.

>>cuttot+(OP)
Who invoked magic in this thread exactly?

>>cuttot+(OP)
I think the point of mine that you're missing (or perhaps disagreeing with implicitly) is that *everything* is fundamentally math. Or, if you like, everything is fundamentally physics, and physics is fundamentally math.

So classes of functions (ANNs) that can approximate our desired function to arbitrary precision are what we should be expecting to be working with.